AnchorFace: An Anchor-based Facial Landmark Detector Across Large Poses

7 Jul 2020  ·  Zixuan Xu, Banghuai Li, Miao Geng, Ye Yuan ·

Facial landmark localization aims to detect the predefined points of human faces, and the topic has been rapidly improved with the recent development of neural network based methods. However, it remains a challenging task when dealing with faces in unconstrained scenarios, especially with large pose variations. In this paper, we target the problem of facial landmark localization across large poses and address this task based on a split-and-aggregate strategy. To split the search space, we propose a set of anchor templates as references for regression, which well addresses the large variations of face poses. Based on the prediction of each anchor template, we propose to aggregate the results, which can reduce the landmark uncertainty due to the large poses. Overall, our proposed approach, named AnchorFace, obtains state-of-the-art results with extremely efficient inference speed on four challenging benchmarks, i.e. AFLW, 300W, Menpo, and WFLW dataset. Code will be available at https://github.com/nothingelse92/AnchorFace.

PDF Abstract

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Facial Landmark Detection 300W AnchorFace NME 3.12 # 1
Face Alignment 300W AnchorFace NME_inter-ocular (%, Full) 3.72 # 18
NME_inter-ocular (%, Common) 3.12 # 19
NME_inter-ocular (%, Challenge) 6.19 # 21
Facial Landmark Detection 300W (Full) AnchorFace Mean NME 3.72 # 2
Face Alignment AFLW-19 AnchorFace NME_diag (%, Full) 1.56 # 8
NME_diag (%, Frontal) 1.38 # 6
Facial Landmark Detection AFLW-Front AnchorFace Mean NME 1.38 # 1
Face Alignment AFLW-Full AnchorFace Mean NME 1.56 # 1
Facial Landmark Detection AFLW-Full AnchorFace Mean NME 1.56 # 1
Mean NME 1.56 # 1
Face Alignment WFLW AnchorFace NME_inter-ocular (%, all) 4.32 # 7
AUC_inter-ocular@0.1 (%, all) 57.69 # 10
FR_inter-ocular@0.1(%, all) 2.96 # 8

Methods


No methods listed for this paper. Add relevant methods here